Experimental hepatocarcinogenesis in the rat has been used as a model to study the cellular and molecular events during neoplastic development. The research is currently focused on defining the possible role of a stem cell compartment in the liver during oncogenesis as well as in the normal liver (see also Project No. Z01CP05262). The hypothesis that rat liver epithelial (RLE) cells may represent a progeny from a common precursor cell for both hepatocytes and bile epithelium has been explored in vitro by examining the effects of different culture media on the differentiation and lineage commitment of RLE cells. We have shown by using differential expression of cytokeratins with molecular weights of 55,000 D and 52,000 D, vimentin, the placental form of glutathione S-transferase and isoenzymes of lactate dehydrogenase as cellular lineage markers, that the RLE cells in chemically defined medium acquire phenotypic traits of bile epithelium. These data further support the notion that RLE cells may be progenitor cells for both bile duct epithelial and hepatocytic cell lineages in adult rat liver. In order to address questions regarding both transplantability and lineage commitments of RLE cells we are currently introducing a retrovirus encoding the E. coli beta-galactosidase (beta-GAL) gene into these cells prior to hepatic transplantation to obtain an easily detectable single cell marking system. We have recently identified cells morphologically similar to rat liver derived oval cells in hyperplastic and preneoplastic human livers as well as in human hepatocellular carcinomas. These putative human oval cells have small basophilic oval nuclei, scanty clear cytoplasm, and cross-reacted with OV-6 (antibody against rat oval cell). Most of these cells express AFP and/or albumin, and are gamma-GT positive.